一类新的纹理谱描述子

在纹理元的基础上提出了一类新的纹理谱描述子,新的纹理谱描述子在3个方面作了改进：将像素的灰度差量化为4个值;量化区间根据纹理对比度自动确定,并保证量化值具有灰度线性不变性;利用相关性弱的8邻域像素构建纹理谱描述子,从而降低了纹理谱维数。定义了基于新的纹理谱描述子的光照、旋转不变性纹理特征。利用该特征对Outex纹理进行光照、旋转不变性分类,分类准确率高于基于局部二值模式的光照、旋转不变性纹理特征。     

基于局部特征多轴旋转不变特性的纹理图像分类算法

采用非量化的局部特征设计出一个稳健的纹理描述符,以便增强旋转和尺度变化时纹理分类的鲁棒性。首先,引入了局部特征的旋转对称性的概念,提出了一种新颖的局部特征来描述纹理的旋转不变特性。为了处理剧烈的旋转、尺度等变化,利用费舍尔向量编码方法对纹理特征量进行多尺度分析,在不增加局部特征维度的同时又能结合尺度信息,由此产生的局部特征对旋转、灰度变化都有较强的鲁棒性。实验结果表明,所提方法的评估结果在许多数据集上都远远超过了现有最优算法,大大提高了纹理分类的精度。     

一种变换域纹理图像特征提取和分割方法

以对偶树复小波变换为基础,提出了一种提取纹理图像变换域统计特征,进而实现图像非监督分割的方法。该方法用Gamma分布和对数正态分布建模对偶树复小波系数的模值,将两种分布的参数综合起来作为像素特征,利用边缘保持的平滑技术（EPNSQ）进行特征平滑,最后使用K－均值聚类方法实现特征分类,得到图像的非监督分割结果。实验结果表明所提取的特征可以有效地表征不同的纹理,基于该特征得到了更为精确的图像分割结果。     

A universal texture segmentation and representation scheme based on ant colony optimization for iris image processing

This paper proposes a novel scheme for texture segmentation and representation based on Ant Colony Optimization (ACO). Texture segmentation and texture characteristic expression are two important areas in image pattern recognition. Nevertheless, until now, how to find an effective way for accomplishing these tasks is still a major challenge in practical applications such as iris image processing. We propose a framework for ACO based image processing methods. Considering the specific characteristics of various tasks, such a framework possesses the flexibility of only defining different criteria for ant behavior correspondingly. By defining different kinds of direction probability and movement difficulty for artificial ants, an ACO based image segmentation algorithm and a texture representation method are then presented for automatic iris image processing. Experimental results demonstrated that the ACO based image processing methods are competitive and quite promising, with excellent effectiveness and practicability especially for images with complex local texture situations.     

Texture segmentation using wavelet transform

Texture analysis such as segmentation and classification plays a vital role in computer vision and pattern recognition and is widely applied to many areas such as industrial automation, bio-medical image processing and remote sensing. This paper describes a novel technique of feature extraction for characterization and segmentation of texture at multiple scales based on block by block comparison of wavelet co-occurrence features. The performance of this segmentation algorithm is superior to traditional single resolution techniques such as texture spectrum, co-occurrences, local linear transforms, etc. The results of the proposed algorithm are found to be satisfactory.     

Retrieval of translated, rotated and scaled color textures

A new method for color texture retrieval using color and edge features is proposed in this study. The proposed method unifies color and edge features rather than simply analyzing only color characteristics. First, the distributions of color and local edge patterns are used to derive a similarity measure for a pair of textures. Then, a retrieval method based on the similarity measure is proposed to retrieve texture images from a database of color textures. Finally, the similarity measure is extended to retrieve texture regions from a database of natural images. Since the proposed feature distributions can resist variations in translation, rotation and scale, our method has the ability to retrieve texture images or regions that change in translation, rotation and/or scale. The effectiveness and practicability of the proposed method have been demonstrated by various experiments.     

利用小变换和特征加权进行纹理分割

为了提高纹理图象分割的边缘准确性和区域一致性以及降低分割错误率,提出了一种基于小波变换的利用特征加权来进行纹理分割的方法。该方法包括特征提取、预分割和后分割3个阶段,其中,特征提取在金字塔结构小小以变换的基础上进行;预分割利用均人矣类算法来对原始图象进行初步的分割;后分割则根据预分割的结果对特征进行加权,然后利用最小距离分类器来实现图象的最后分割。与传统的方法相比,该方法在分割错误率、边缘准确性以及区域一致性等方面均有明显的改善。     

抑制纹理信息的偏置场变分图像分割模型

偏置场变分水平集图像分割模型利用原始图像的局部灰度信息,可以对灰度不均匀图像进行有效的分割,但当灰度图像中存在纹理时,分割效果往往很差。针对这一问题,提出抑制纹理信息的偏置场变分水平集图像分割模型。利用一种基于纹理几何结构的纹理描述符描述图像中不同的纹理区域,使得不同纹理区域对比更加明显,相同纹理区域更加平滑,通过抑制纹理信息使后续的图像分割在纹理部分的错分大大减少。实验结果表明,相比偏置场变分模型,所提模型对自然及人工合成纹理图像均获得更好的分割结果。     

A clustering method based on multidimensional texture analysis

Considering the analogy between image segmentation and cluster analysis, the aim of this paper is to adapt statistical texture measures to describe the spatial distribution of multidimensional observations. The main idea is to consider the cluster cores as domains characterized by their specific textures in the data space. The distribution of the data points is first described as a multidimensional histogram defined on a multidimensional regular array of sampling points. In order to evaluate locally a multidimensional texture, a co-occurrence matrix is introduced, which characterizes the local distribution of the data points in the multidimensional data space. Several local texture features can be computed from this co-occurrence matrix, which accumulates spatial and statistical information on the data distribution in the neighborhoods of the sampling points. Texture features are selected according to their ability to discriminate different distributions of data points. The sampling points where the local underlying texture is evaluated are categorized into different texture classes. The points assigned to these classes tend to form connected components in the data space, which are considered as the cores of the clusters.     

基于小波分形特征提取的图象分割方法

提出了一种基于小波分解和分形纹理特征计算的图象分割方法,首先考虑对图象进行小波变换,然后对不同通道的子图象提取纹理的分形特征和能量特征,最后用直方图阈值分割方法实现图象的分割,实验表明,该方法对模拟纹理图象以及多少谱遥感图象的分割都取得了满意的效果。     

无监督纹理遥感图像分形分割仿真

针对无监督纹理遥感图像分形分割时,仅利用光谱信息,没有充分利用遥感图像中的纹理信息造成过分割或是欠分割的问题,提出结合光谱信息和纹理信息的无监督遥感图像分形分割方法。方法首先借助于双边滤波算法建立双边滤波模型,对无监督遥感图像进行平滑处理,利用边缘检测算子获取无监督遥感图像的纹理梯度和光谱梯度,保留遥感图像的纹理信息。将图像纹理梯度和光谱梯度的内积范数作为K-means算法的距离测度,针对K-means算法易陷入局部寻优问题,将遗传算法与K-means算法相结合实现全局寻优,完成无监督纹理遥感图像分形分割。通过在高分辨率遥感图像上的分割对比实验,证明上述方法能充分利用遥感图像的光谱和纹理信息,分割效果良好。     

Image segmentation using local spectral histograms and linear regression

We present a novel method for segmenting images with texture and nontexture regions. Local spectral histograms are feature vectors consisting of histograms of chosen filter responses, which capture both texture and nontexture information. Based on the observation that the local spectral histogram of a pixel location can be approximated through a linear combination of the representative features weighted by the area coverage of each feature, we formulate the segmentation problem as a multivariate linear regression, where the solution is obtained by least squares estimation. Moreover, we propose an algorithm to automatically identify representative features corresponding to different homogeneous regions, and show that the number of representative features can be determined by examining the effective rank of a feature matrix. We present segmentation results on different types of images, and our comparison with other methods shows that the proposed method gives more accurate results.